Figure 1
Graph representing the data from Cavanagh (1972). The quantity of items through which one can cycle per second is represented on the x-axis, and the memory span on the y-axis. The figures represent the different materials included in the meta-analysis.
Figure 2
Illustration of the memory span and memory search tests. In both of these tests, participants are first presented with a memory list. In this example, participants have to remember the location of a series of squares in a matrix. In the memory span task, participants subsequently have to click on the correct square in each black matrix, respecting the order of items in the memory list. In the memory search task, participants are presented with a single stimulus, and have to decide whether or not this stimulus was present in the memory list. Different materials will be used in the experiment, all with comparable recall procedures.
Table 1
Description of different verbal memoranda used in the pilot experiments.
| Verbal memoranda | |
|---|---|
| Digits | Randomly sampled from 0 to 9, presented in uppercase Times New Roman 48 font. |
| Letters | Randomly sampled from 19 consonants (excluding Y and W), presented in uppercase Times New Roman 48 font. |
| Words | Randomly sampled from a set of 312 common French one-syllable words (4 to 6 letters, excluding special characters), based on a French text database (http://www.lexique.org/, developed by Université Savoie Mont Blanc). The items were printed in lowercase Times New Roman 24 font. |
| Pseudowords | Randomly sampled from a selected set of 312 French one-syllable pseudowords (4 to 6 letters, and excluding special characters), matched to the words in terms of frequency of the bigrams involved. The items were printed in lowercase Times New Roman 24 font. |
Table 2
Description of different visuospatial memoranda used in the pilot experiments.
| Visuospatial memoranda | ||
|---|---|---|
 | Locations in a matrix | Randomly sampled from 16 possible locations in a 4 × 4 matrix consisting of 25 mm × 25 mm squares. |
 | Irregular locations | Randomly sampled from 16 possible locations indicated by 25 mm × 25 mm squares that appeared on the screen in a fixed and irregular pattern. |
 | Arrows | Randomly sampled from 16 possible arrows, that radiated outwards from the center of the screen and differed in length (2.5 cm or 5 cm) and angle (0°, 45°, 90°, 135°, 180°, 225°, 270°, 315°) |
 | Ball movements | Randomly sampled from 16 different movements of a disk (diameter 1,5 cm) within a square box spanning 20 cm on each side. The disk could move on one of 16 possible paths. The possible starting positions corresponded to the corners and the midpoints of the sides of the square. The disk could then move vertically, horizontally or diagonally to the opposite side of the square, in both directions. |
 | Lollypops | Randomly sampled from 16 different lollypops, pointing outwards from the center of the screen, and differing by their angle (from 0° to 337.5° by increments of 22.5°). |
 | Squares | Randomly sampled from 16 different squares, appearing in one of four positions in a 2 × 2 matrix (this matrix spanned 26 cm on each side), and differing in size (sides of 1.5 cm, 3 cm, 5 cm, 8 cm). |
 | Ice-cream cones | Randomly sampled from 16 different cones measuring 2.5 cm in length, appearing in one of four positions in a 2 × 2 matrix (this matrix spanned 16 cm on each side), and differing in orientation (45°, 135°, 225°, 315°). |
Table 3
Mean span and standard deviation (between parentheses) for the different materials tested in both pilot studies.
| Memoranda | Span (SD) |
|---|---|
| Verbal | |
| Pseudo-words | 2.06(0.37) |
| Words | 3.39(0.65) |
| Letters | 4.81(1.04) |
| Digits | 5.44(1.23) |
| Visuospatial | |
| Lollypops | 2.61(0.44) |
| Ball movements | 2.79(0.48) |
| Ice-cream cones | 2.88(0.7) |
| Arrows | 3.14(0.72) |
| Irregular locations | 3.19(0.72) |
| Squares | 3.3(0.66) |
| Locations in a matrix | 3.31(0.67) |
Figure 3
Expected relation between the measured memory span for the selected materials (y-axis), and the memory search speed (items/s) on the y-axis. The circles represent verbal memoranda and the triangles represent visuospatial memoranda.
Table 4
Information concerning different materials; including average memory span, average RT (in ms) for probe-present trials, slope relating RT to list length, inverse of slope (items per second), accuracy on probe-present trials, and slope relating accuracy to list length.
| Capacity test | Speed test | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Span M | Span SD | RT M | RT slope (ms/item) | RT slope SD | Inv RT slope (items/s) | Acc M | Acc slope (%/item) | ||
| Verbal | Letters | 5.26 | 0.87 | 852 | 37 | 58 | 27 | 93% | –1.8% |
| Words | 3.84 | 0.64 | 840 | 19 | 51 | 53 | 94% | –0.3% | |
| Pseudowords | 2.08 | 0.38 | 862 | 38 | 50 | 26 | 95% | –1.9% | |
| Visuospatial | Matrix | 3.56 | 0.59 | 980 | 28 | 56 | 36 | 93% | –1.6% |
| Icecream | 2.56 | 0.60 | 998 | 74 | 49 | 14 | 89% | –3.0% | |
| Lollypop | 2.44 | 0.62 | 1064 | 31 | 66 | 32 | 80% | –4.2% | |
Table 5
Summary of evidence from analysis on different subsets of participants. BFs in favor of a positive relation between memory speed and span are indicated in green, BFs in favor of the absence of a positive relation are indicated in red, and anecdotal evidence is indicated in grey.
Figure 4
Observed relation between the measured memory span for the six materials (y-axis), and the inverse of the measured memory search slopes (x-axis) in the planned restricted sample (N = 9). The circles represent verbal memoranda and the triangles represent visuospatial memoranda.
Figure 5
Number of occurrences of negative search slopes for letters and matrices over blocks within the first session (1, 2, 3) and the second session (4, 5, 6).